Does Phase Distortion/Shift Matter in Audio? (no*)
HTML-код
- Опубликовано: 22 авг 2024
- A look at audibility of phase shift/distortion in audio and its practical implication with headphone and speaker listening. Research studies are shown as well as measurements to show the impact of phase deviations.
Discussion thread on Audio Science Review: www.audioscien...
Perceptual Effects of Room Reflections: www.audioscien...
Audeze LCD-24 Headphone Review: www.audioscien...
Revel C52 Speaker Review: www.audioscien...
Dr. Floyd Toole's Book: Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms
www.amazon.com...
On the Audibility of Midrange Phase Distortion in Audio Systems:
www.aes.org/e-...
Clearly Paul must clarify his statements, if nothing else, to save phase
Maybe.... someone can explain what a phase shift is?
That right there is really funny. :)
I hope that this doesn’t turn into a phase off.
Phase shift happens …
@@kristiantizzard7796
I certainly hope so
Paul McGowen - the most confused & least informed CEO of an Audio company - I’m sure some of their products are well engineered ( by other people ) but the avalanche of Bull emanating from Paul is staggering!!!!
Indeed they do produce a mix of products, some of which are good, but others are not as they are led by folklore than proper science and engineering. Their DSD DAC is one example where it performs horribly yet he and others there think it sound good to their ears, while listening with their eyes....
See: www.audiosciencereview.com/forum/index.php?threads/review-and-measurements-of-ps-audio-perfectwave-directstream-dac.9100/
Chris Brunhaver sounds like a top notch engineer. So, PS Audio speakers should be excellent.
@@Rene_Christensen I agree 100% - unfortunately most people are clueless regarding acoustics - it's complexity & nuance - including the vast majority of the "hi-end" fraternity - starting with the clueless dealers - it's embarrassing.... but hey, not as crazy as volunteering to be a guinea pig .....
Paul pretty much never answers any question straightforward
Mumbo jumbo sells products. "If you can't dazzle them with brilliance, baffle them with bullshit." He wants people buying his products, the truth isn't important and in most cases is a hindrance to that goal.
Not unless he is GENUINELY confabulated. Which may indeed be the case for all with experience or academics. Who know .... science and engineering textbooks 100 years from now my totally support McGowan and left Majidimehr steppin' on his own dick.
10 min. mark needed
His answers are base more in philosophy that than science!
Most of these "audiophile" people don't, especially if they have a business.
The best you'll get is some wishy washy that isn't a direct answer, or if you get a direct answer, it will be "It just is better"
Amir is in a good phase!
Hehe. :) Nice one.
All the best snake oil takes a scientific fact as its basis and projects it into the realm of fantasy. Paul's fan club consists largely of the people who spent their science classes staring out of the window. I wouldn't mind if they were the only ones falling for it, but too many of them take his words as gospel and spread the myths and legends to newbies across the internet. Thank you, Amir, for standing up for science and truth!
Totally agree.
Well said and my pleasure. :)
Amen!
@@AudioScienceReview so how come you not done 5 videos between the last 5 weeks now. did ps paul get angry
Sheep everywhere
Paul is mostly a salesman that randomly throws around technical terms to add credence to his sales pitch.
Slimy old creep in a "nice friendly uncle" mask. Can't stand the man. At least he's not as ugly as Guttenberg though 😄
Nicely explained Amir. In the RF world we use delay and reflections to measure wire lengths and impedance discontinuities. I often tell people, a good understanding of RF principals eases the understanding of audio principals.
Great point.
absolutely
I agree as well, although it's extremely ironic that you'll find quite some RF engineers believing in such "interesting" stories.
Your content really fills an important void in audio education. I applaud you Sir
Ah, you are very kind. I have to get motivated to do videos and comments like this provide that motivation.
@@AudioScienceReview as someone that doesn't like audiophile mumbo jumbo, you're the best teacher in acoustics that I've encountered. Great presentation, clear info, easy to understand yet doesn't skim over details, and, no bs. You're really a gem on RUclips Sir!
Paul’s followers need to realise he is a salesman with superficial technical knowledge. I have watched many of his videos where he demonstrates a total misunderstanding of engineering. Take a look at his videos on audio tuned fuses, elevating speaker cables, different speaker cable design for bass v treble, audible effects of speaker cables, audible differences between usb cables, white board drawings of phase shift, explanation of DSD v PCM,, power regenerators etc. etc. It is alarming that so many people are fooled by snake oil, and not just in audio. This indicates a serious inadequacy in the education system. Our kids are not being taught basic science and how to think critically.
You make me realize it would take a lifetime to produce videos to counter the misconceptions he is handing out. If I didn't have other work to do, I could take it on but I do. Still, if people point out specific videos, I will produce some as I did.
@@AudioScienceReview ruclips.net/video/kFlnQ1chBno/видео.html
This video says it all. Apart from the pure BS of audio tuned fuses Paul does not know why some fuses are sand filled. He clearly has no formal qualifications in electrical engineering.
This is what he says about himself: www.linkedin.com/in/paul-mcgowan-73101611/
Education: Valencia High School
"I am a self-taught analog electronics engineer and have designed most of the analog product built by the company."
A "self-taught engineer", whatever it means. I think he was a enthusiastic electronics hobbyist, and has become a self-taught designer and CEO. We can't expect well based scientific knowledge and scientific temper from him.
@@AudioScienceReview I really hope that you can help "re-educate" some of these snake-oil-polluted audiofools, Amir. Most of them are VERY resistant to accept the truth, after swallowing the BS from so-called "experts" for so many years. We engineers & genuine hifi enthusiasts appreciate your efforts very much though, even if just to bolster our own firmly-held beliefs & knowledge (and give us some ammo to fire back at the crazy audiofools!)
I enjoy watching RUclips videos about audio, for advice and entertainment. I am also aware that most presenters have an end game which then will biases there opinion. So this usually colors their comments. So that's what I personally value Amir's advice highly, cause for myself I do not believe He has any "end game" except maybe knowledge.
That's very kind of you.
But not true, he has loads of bias, opinions and conjecture. Even when he says what Paul said is true he then wanders off point conjecturing about his rationale. His schtick is criticizing others and rolling out his biases. So do phase shifts matter? Sub woofer set up suggests they do… even your own graphs suggests that phase matters, no matter where they come from. Speaker performance is definitely impacted by phase. So is amp phase in audible? No it’s audible because it causes speaker phase shift. Your presentation is simply like room reflections, to mixed… way too speculative. You revert to personal comments and claim his stories. I suspect half of what you say is equally speculative as Paul’s is. Unfortunately your graphs are equally useless, incomplete self contradictory. Oh and if you don’t understand phase shift you’re useless too? You are here primarily to prove yourself right rather than educate, rather develop new knowledge. Like you said, you rant too much. Such a shame. You and Dennis at GR Research are actually very similar in cynical outlook. Must be a frustrating way to live.
My phase has shifted.
Actually, PS audio was the worst amp I ever owned. Had a 5v DC leak and no filtering capacitors.
No filtering capacitors? Maybe that's so he can sell you an external power conditioner.
Which model was it?
@@0x07AF I'd have to find it. It's been a while. As far as I remember the power supply was called direct coupling or something like that.
@@snowblow1984 Ahh.. Direct coupled sounds like it was designed to work with no capacitors in the signal path...for an amplifier though. Sometimes the circuits used to compensate and adjust the signal so it works without caps can cause worse phase shifts than the capacitors would have. DC servos in the feedback of the output amplifiers are known to do that, and they're especially bad if poorly designed.
I have no idea why they would also avoid capacitors in the power supply too, to me that would border on crazy-town.
Was it an older product? Thanks for replying!
@@0x07AF It was an older product. I am trying to find at list a picture of it but it's been a few years.
Thanks Amir. You and Audioholics saved me from the rabbit hole of cables. The ones I have are fine. The only slightly expensive cables I tried where professional ones. Mogami, Gotham and Canare. If they are used to capture music recording, then they are fit for playback at home. No need for exotic marketed "audiophile" stuff.
My pleasure. Cable business was created to make money for retails first and foremost. When cable reps come to pitch you on them, the first line is, "this cable will give you 65% margin -- how are you going to get that from selling electronics?" Answer is that you can't. This is why you can't leave BestBuy or other retailers without buying a cable to go with your hardware purchase.
@@AudioScienceReview , when I worked at Best Buy, our cost on audio/video cables was roughly 10-25% of the MSRP.
I don't know much about audio and there was a time I watched Paul's videos. Then, one time he touched the point I know very deeply - since then I deleted his channel from my favourites ;-)
And not only his btw.
But still, he is ... charming ;-)
He is a salesman above all.
@@TobyIKanoby being a salesman is NOT a problem but saying BS to the people who are asking his for serious advice is... I am lost for words
@@zyghom sure, being a salesman is not a problem, I don't think I said it is. But when I am watching his videos, that's what I see although he doesn't really disclose it well, on the contrary, and it is always healthy to be skeptical about anything a salesman says. Basically, I am saying the same thing as you are.
@Mike Glavin yeah, it's a little bit creepy
In audio synthesizer world, a supersaw synth is bunch of saw waves (oscillators that generate saw waves to be more precise) on top of each other. You can get different sounds by different count of the saw waves and spreading each saw wave very slightly in pitch and stereo pan. If the phase of each saw wave is randomized and each time you hit the same note, the waveform is completely different, but the perceived sound is same.
Good point.
There are also synths that use phase distortion/modulation for synthesis en.m.wikipedia.org/wiki/Phase_distortion_synthesis
I agree with some of what you said.
It's important to understand you will always perceive best (hear) what is loudest. If a speaker has inherent phase issues, compared to a speaker with flat phase, then put in the same situation (placement in a room), the speaker with correct phase will produce a "better" sound.
Also...speakers can be used outside where reflections and late arrivals are not an issue, in this situation phase is extremely important to the impact and "feel" of the sound.
PHASE DOES MATTER
When I saw Paul's video I thought to myself, one of those science audio guys is going to rant it... Bingo! Thank you
At your service!!! :)
Pretty much the first thing you learn when studying audio compression is "throw away the phase". It halves the data in a single move. If phase had any significant effect on human perception of audio, well compressed music would not be so very hard to tell from the original. That said, a blanket statement that phase doesn't matter is misleading. At higher frequencies it certainly doesn't matter, but at bass frequencies, where the wavelength is long, it often can.
Let's phase it, it doesn't matter.
Another great presentation Amir! You know, it’s interesting. The National Research Council (NRC) in Canada funded a lot of the acoustic research Floyd Toole has done. And a lot of really great products were born in NRC research labs - Energy and PSB were/are beneficiaries of Canadian government funded research. Mr. Toole’s book, Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms is an essential reference for anyone into audio.
I think if one guy needs to read that book is Paul McGowan. In one of his videos, if I remember right, he seemed to indicate he didn't know who Sean Olive/Floyd Toole were! And here they are designing speakers? It is like wanting to be a physicist yet not having heard of Einstein. The degree to which these people ignore science is incredible. Everything is based on lay understanding of audio. He thinks phase is bad in his mind, so it must be.
@@AudioScienceReview nah...he's all about PT Barnum's bromide.
This was an interesting video but it's incomplete. What Amir is showing (i.e. when a signal's phase is changed at higher frequency due to effects of a filter) is referred to as group delay. Group delay simply means that not all frequencies within a given bandwidth pass through the filter in the same amount of time. The end result (in this situation) is that you wind up with the low frequencies having little or no phase shift and the high frequencies with substantial phase shift. What are the implications of this? Say you have a musical instrument, such as a guitar, with a lot of harmonics that give it a "fairly wide" bandwidth. If you pass this through a band limited channel (i.e. filter) then you are going to phase shift the upper frequencies of the instrument while leaving the lower frequencies unchanged. This is not a simple situation of applying an equal phase shift to all frequencies within the signal (that would certainly be inaudible I would assume). The question then becomes is group delay audible? I don't know the answer and it may be such a minor effect as to be washed out by the room reflections that Amir addresses. I'm simply pointing out this video is somewhat oversimplified as to the complete picture of what is going on with phase change due to a band limited channel.
Thank you for the excellent review on this topic. For the first time I don't 100% agree. I did some experiments myself with IIR and FIR filters on a 3 way system. The FIR filter has other than the IRR no phase shift. In near feeld I can hear the difference in my refletive room, particularly with string instruments like piano. With increasing distance the room dominated. I blind tested it and it makes a difference. I encourage you to test it for yourself. Subjectivity it changes the illusion of the room of the recording.
U should contact Dr F.Toole and let him know asap
I liked your video I thought you had some pretty good information here. HOWEVER... haha. You say that the human perception of phase from an audio source is flawed to begin with and you give the example if you were sitting next to me and talking, your voice would be perceived as distortion-free despite all the parasitic elements of the room and such since the brain automatically compensates. That part is all good. It's crazy, the relationship between your brain and ears. I am an audio engineer and I always say knowing your room is #1 above anything else. I would take a job mixing in my studio on my equipment any day over the best studio on the best equipment. Moving the best equipment in here would be ideal but hey.... we're working on that. Ironically, this is also what makes your statement about listening to your voice irrelevant. When listening to a recording, the brain actually has to that assessment twice because there was another separate set of ears involved in hearing that sound indiscriminately and I'm talking about the microphones. They also picked up the signal, complete with reflections, phase distortion, the whole nine. So the reflection thing you said. double that. But also, the brains ability to decipher the mess and if it didn't we'd go crazy from all the excess. Double that too. The brain can do it without even breaking a sweat. That's why literally anyone who can hear, if they are sitting in a closet listening to a sample of audio recorded at the Taj Mahal they can easily tell you that the reverb they are hearing is definitely not a result of the closet. And with a little ear training someone can tell you with certainty that it's the recording that's too bassy or too trebly, not their playback system. With even more ear training someone can tell you that the bass guitar and kick drum in a recording could have more clarity but they are being masked by the excessive 250 Hz in the guitar track. If you don't already see where I'm going with this, it's not as far fetched as you may think for a significant number of people to be able to hear phase distortion in audio playback despite real-world convolutions. They might not even know what phase distortion is but they can hear it. That's where electrical engineering comes in to play with the audio world. If you were an audio electronics engineer without any ear training (not recommended) wouldn't you at least design with those trained ears in mind? If you cared about audio quality, you sure as hell would. Whether anyone actually cares about that or not is another thing altogether but you can bet they all say they do haha. Cheers
I'm glad you have addressed this snake oil salesman, what baffles me is so many people so gullible to just believe anything he says and buy multi k$ equipment to solve problems that don't exist and then claim they hear a night and day difference...
Paul McGowan is to audiophiles as Mr. Rogers is to children. It doesn't matter what he says. It's the soothing way he says it.
I am happy to do more of them if you message me on Audio Science Review as the person did that trigged this response. I don't watch Paul's videos often enough to catch these things. We definitely need to set the record straight on many of these topics.
@@cowboyflipflopped you beat me to the Mr. Rogers comment!🤗
I think it's important to make the distinction between monaural phase distortion and binaural phase distortion. It has been shown again and again, ie this video, that the human ear is not sensitive to monaural phase. However, when interaural phase is distorted, the inherent positioning of audio objects in the acoustic scene is distorted as well. This is what I believe a lot of audophile reviewers refer to when they talk about soundstage. I think it would be interesting to include some channel crosstalk or channel phase distortion measurements in your reviews because they would be the measurements that would dictate the accuracy of the transmitted soundstage.
I feel like with soundstage there is still so much snake-oil being thrown around because some of it has to do with incorrect frequency response, but to my knowledge when looking at the accuracy of the soundstage, the only thing that should matter is the interaural phase distortion.
Changes in transient response due to phase shift from filtering can be clearly audible, though, but my only experience with that comes from equalization during production (recording/mixing/mastering) rather than as an end-user.
It is audible as night and day IME as well. I am unclear if these audioscience guys are deaf as fk or just petty liars.
Thanks Amir, as usually great info with proofs and explanations. I'm adding the video to favorites list for future reference
My pleasure. Thanks for watching.
If everything we heard in high end audio were true, we never would have heard it to begin with
Thanks for another great video Amir. I've often felt like there's an aura of mysticism around Paul McGowan's technical explanations. That's usually a sign of disinformation being spread.
My pleasure. I really want to know if he really believes this stuff or just says them because he knows it works as a sales tactic.
Thank you Amir,
I so enjoy listening to the way you explain audio in a technical but understandable way.
I really appreciate your wide knowledge and experience in all things audio. I look forward to your future insights and explanations .
Thank you for this video. It has helped me out with a dilemma of my own. I use Sonarworks Reference 4, which offers multiple filtering options for corrective EQ. You can choose either Zero Latency, which introduces phase shift, or Linear Phase, which introduces pre-ringing artifacts. I was worried about what effect the phase shift could have. But now I know that it's not an audible problem at all. Yay! I can have my audio cake and eat it, too.
Amir, I really wish you were my DSP professor in college. I would have spent more time on that subject and learnt the concepts more in depth. It is really a joy watching your videos and the way you explain some of the harder concepts so well.
I know this reply is very late, but if you want to brush up on DSP, there is an EXCELLENT free textbook available, "The Scientist and Engineer's Guide to Digital Signal Processing" by Steven W. Smith. It has enough technical depth to make for a very good resource, but it is also about as easy-to-read as a book on the subject could be. A quick search for the title should make it easy to find :)
@@MikeyAntonakakis thanks for the suggestion. its good to get to basics time to time.
What about Ethernet cables? Cat7 is 600MHz, Cat8 2GHz. Not infinite bandwidth, but higher the better, the reasoning being the following: "At least three things matter - the clock, noise and bandwidth. In the image of a perfect square wave, the horizontal axis is time and the vertical axis is voltage. We will assume the clock is perfect - ie. the vertical signal lines occur at perfectly spaced intervals (the bit rate). When the signal is representing a binary 0, it is at 0v. When the signal is representing a binary 1, it is at 1v. And we will assume that the receiver of this signal decides that the transition between a 0 and a 1 has occurred when the signal rises through the 0.5v level, and that a 1 has transitioned to a 0 when the signal falls through the 0.5v level. Now imagine that there is noise added to the signal. If the frequency of the noise is below the bitrate then this perfect square wave swims on top of a longer and smoother wave. The interesting point is that the timing between the data transitions (where those vertical lines pass through 0.5v) is unchanged. So no problem, yet. If the frequency of the noise is above the bitrate then the horizontal lines get fuzzy. And if we combine the low frequency noise with the high frequency noise the effect is combined. Again, the interesting point to note is that the timing between the data transitions (where those vertical lines pass through 0.5v) is unchanged provided the noise is not extremely high. So, again, no problem. Noise, on its own (as long as the deviations caused are materially below 0.5v) is not a problem. The reason it is not a problem is those vertical lines, because noise does not change the space between them.
Now imagine there is no noise. Zero noise is impossible, but something else that is impossible is the vertical line on the square wave, since it requires infinite bandwidth. The vertical lines imply the signal can achieve 0v and 1v in more or less the same instant. Whatever tools we have to transmit a signal, the demands of high bit-rate signals are way beyond what the available tools can deliver. Think about how your analog cables can mess with sound up to around 20kHz, and then think about the enormously wider frequency range required of a digital cable (and, optical cables just have a different set of problems, mainly related to reflections). The higher the bit rate the harder it gets. When we allow for constrained bandwidth, instead of transitions being instantaneous, the signal goes up a slope when transitioning from 0v to 1v, and down a slope when transitioning from 1v to 0v. If the bandwidth was the same as the bitrate then the signal would be a sine wave. To reasonably square out the signal you need to add several harmonics of the bitrate (say 7 or more) above the bitrate, and that is a lot of bandwidth - even more for higher bit rate signals. By adding harmonics, the sine wave begins to square out. Interestingly, in both of these constrained-bandwidth examples, the transitions through 0.5v are still perfectly spaced - even with the sine wave. So still no problem.
But as I mentioned, a higher bitrate signal (if you think high bitrate files must always sound better) requires even more bandwidth to square out the wave, and so in a system that has a finite limit on bandwidth, a lower bitrate signal will be more accurately represented than a high bitrate signal. On top of that, if you ask anything in a music server to work faster, it will work with less precision and this is a key trade-off to be aware of when you assume higher bit rates must be better, just because the numbers are bigger. These examples only allow us to conclude that there is no problem if we can achieve zero noise or infinite bandwidth. But each of those goals is unattainable, and the problem becomes apparent when there is both noise and constrained bandwidth. So what happens if we add a low frequency noise component to a frequency-constrained digital audio signal? All of sudden, the 0.5v points are shifted right or left by the addition of the low frequency noise that lifts or drops the signal between bits. Shifting the slopes up or down shifts the 0.5v points left or right. The greater the amplitude of the noise, and the greater the bandwidth constraint, the greater is the effect on timing (jitter).
Now if we add high frequency noise to a frequency-constrained signal you can see that the transition timing at precisely 0.5v is now hard to discern for any digital receiver. If the signal is vertical at the transition then noise does not affect it. But as soon as the transition is not vertical then noise changes the transition point. It is the combination of constrained bandwidth and noise that inevitably creates jitter (variation in data transition timing), regardless of how great the clock is."
Your basic understanding of how digital signals work, namely ethernet, makes the whole post invalid. For example, voltage is +-2.5V over a twisted pair.
It is true that lowering the bandwidth reduces the slope of zero crossing and with it, increase the impact that noise has in creating jitter. It is also true that the receiver has phase locked loop and filters such noise and captures the clock. So as long as your eye pattern is wide enough, you are good to go. Having more bandwidth doesn't buy you anything. It is like your dishes. Do you wash them once or 100 times? They may be cleaner if you wash them 100 times but you don't do that because it has a cost to it (time and money). Same with bandwidth. It doesn't always come for free although in a linear amp, you can easily get more than 20 kHz.
Go ahead and change your cat 5 cable to cat 6 on your short connection from a switch to your computer. Does it improve bandwidth? It will not at all. Only over very long distances where the cable filtering becomes significant does it matter. And even there, it has to hit that cliff where data recovery suffers. With the checksums in each packet, some amount of data error is accept regardless.
Anyway, this is way outside of the scope of this video. Here we are talking about phase, not bandwidth. As I mentioned in the video, some extra headroom is fine to have.
@@AudioScienceReview As the matter of fact, here is what MSB is saying re Cat5/6 and their Analog DAC:
"
MSB Renderer optimization
After testing by us and then many other dealers and customers, we have settled on a scheme for ethernet filtering.
It requires:
DJM Electronics GigafoilV4 www.djmelectronics.com/gigafoilv4-inline-ethernet-filter.html (not sure about current pricing but the retail was around $550)
The iFi 5 volt power supply www.amazon.com/iPower-Supply-International-Travel-Adapters/dp/B01GNNXP0Y $49
The latest and better power supply solution according to a consensus is the Keces P3 www.kecesaudio.com
We recommend, DO NOT use CAT 7 going to the input of the GigafoilV4, and DO USE CAT7 between the GigafoilV4 and the MSB Render input. This is more about the CAT 7 being a ground and where you want, and do not want that ground, rather than CAT 7 having ultra high bandwidth which we don’t need for our purposes. Again, please adhere to the use of CAT 6 and CAT 7 just described. (note: people seem to want to use a CAT 7 going into the Gigafoil, DO NOT, DO NOT, DO NOT use an audiophile ethernet cable into the Gigafoil. Use a plain Jane CAT 6 going in !) listen for yourself and compare if you are tempted to use an audiophile ethernet cable into the Gigafoil, you have been warned!!!).
We have had a nice improvement using the Audioquest ethernet cables which are CAT7. I also have good reports about the Wireworld cables including the Platinum 7.
There are also good reports about using the Keces P8 19 volt power supply on the Roon Nucleus and any Roon core (such as the Intel Nuc or a laptop), that runs on 19 volts. www.kecesaudio.com"
So if we change Cat5 to 6 (as a headroom as you said) and Cat6 to 7, MSB recommends router to ($550) switch cable be cheap Cat6 and switch to Analog DAC Ethernet input be Cat7.
The only thing is that you already debunked $640 ER switch and there is no reason this Gigafoil $550 switch is any better that $20 switch. Some people claim ER is not a filter while others claim it is (among other things). Your measurement showed it is not a filter (otherwise it would show a different spectra than $20 switch). I personally think that by pricing ER at $640 is a proof it is a snake oil. When one opens it and compares with designs of some amplifiers, it is clear like a blue sky it is a hoax. Even $100 would be too much, even if it does what it claims. I paid $100 for Roku 3 and it rocks. They could have selected the price of $300 and maybe nobody would suspect anything. But $640?!?! Insanity. And even after debunking it loud and clear, there are whole movements out there despising the debunking?! And THAT is what Uptone counts on: on unprecedented stupidity of audiofools. If they priced it at $1000, nope, too many digits for a few resistors and capacitors. $300, nope. Audiofools would not consider it audiofool worth. And that's how they came up with $640 price. Audiofool price.
This is lowkey one of the funniest audio enthusiast channels on RUclips
As an audio engineer I feel compelled to point out that phase always matters. It isn't an issue until it is one. The question is what phase are we referring to specifically and how much does it matter in terms of audibility.
Although you're absolutely correct, everything phase-related in the audio engineering world is still differential phase as described in the video. It's only an issue if you have two very similar signals, but when solo'd, you wouldn't hear a thing which is mainly what's being discussed. I personally can (just about) hear the effect of an all-pass filter, especially on transients with its cutoff frequency specifically set so it creates the most audible difference, but it's incredibly subtle and I was going out of my way to try and listen to it, switching back and forth between the filter being on and off. No one listening to music in the real world ever has the luxury of us audio engineers to A/B different versions of their music with different phase reponses to ever tell the difference anyway. Heck, I actually preferred the sound of the kick drum with an all-pass at 150Hz :)
I understand that an amp that is phase correct and has a bandwidth of up to 1 MHz isn’t really needed for human ear and brain, but:
I have questions:
when phase distortion is irrelevant, ie phase distorting filters etc may be applied, how does that keep vital phase relationships between right and left channels untouched - if we want this to work also for complex music signals where left and right channel components might also be different in frequency & timing but constitute a few sound events the brain puts together? Is the correct way to say transformation involving constant or linear phase errors (as fct of frequency, identical for left & right ch) is ok for the brain… but perhaps more complex phase manipulation might not be necessarily be?
How can percieved timing relationship between different parts of one instrumental (think of snare drum) sound coming from one single channel be maintained if phase was irrelevant?
And - if phase was irrelevant - what is the trouble in recreating an accurate signal from digital domain? I always thought its to keep alias out of audible band, but that introduces phase errors in impulses - wrong? Why all the phase filters in some DACs, which apparently do not much but still something?
And why do phase neutral active speakers with dsp´ed x-over sound more real w.r.t. impulses? And what are the main ingredients for excellent PRaT? Phase obviously not.. or isn‘t it to some degree?
Oh one correction: the ear & brain certainly disentangle the sonic mess in an echoy room by being able to single out the first (incident) wave of one sonic event (a sonic event is what the brain makes out of it) - ah! How to tell if the brain and ear would entirely throw away any time (or phase) information? Is time important but phase irrelevant?
Its not so black & white as it all sounds
Thank you for including detailed headphone, speaker examples, and published papers on the audibility of Midrange Phase Distortion.
By the time I learned enough about audio to understand all these measurements, I realized I couldn't hear them any more!
So what about this: The initial direct sound can indeed serve as a sort of “reference” for our auditory system, helping it to anchor or prioritize the sound’s timbre, pitch, and spatial cues before the reflections add their complex overlay.
This process is related to the precedence effect or Haas effect, where the brain prioritizes the first arrival of sound for directionality and spatial localization, allowing us to focus on the direct sound and somewhat ignore the early reflections and reverberations. This mechanism helps in understanding speech in noisy environments and also in maintaining the clarity of music in reverberant spaces. - Doesn’t this mean the absolute phase of the system is important and aren’t transients and timing also severely affected by bandwidth?
I am not an engineer, but an amateur musician that has played and conducted a chamber orchestra. Your argument is correct and what you mention happens in a concert hall. There is direct and reflected sound and what you hear is a mix of both. And when the concert is in a highly resonant venue, like a church, you will have an increased amount of reflected sound.
Thanks for chiming in. At first when I read your intro I thought you were going to disagree and was relieved that you did not. :)
BTW, I am acquiring a new pair of amplifiers. I am seriously considering a pair of Benchmarks AHB2 in bridged mode and use NL4 soeaker cables because I biwire the speakers. I tead your Benchmark review and I was impressed.
Paul always says a LOT, to actually say very little. Tiring. And I would like to see the PS Audio Power conditioners compared with 'lampcord' at the speaker output. I doubt there is even 0.1db difference across the spectrum. Only diff is in your bank account. I have a friend who blew nearly 10K on a Shunyata 'conditioner' and associated cables and such nonsense. AB test? Nope. But of course, it makes all the difference in the world!!! So. We are to believe that audio designers are ignorant, and the cable and power conditioner folks are the magical wizards. Hahahahaa.
I have one of his older power conditioners but keep forgetting to do a test on them. I will do so soon. :)
How about using DSP for crossovers or any EQ for that matter. Do they introduce a phase shift.
Analog, and most DSP based, frequency filtering introduces phase shift. Zero phase filtering is possible thanks to the power of math but only via computation and has many consequences, including changing the frequency response of the output (blurs transients). Bottom line: can't touch the frequency content of a signal without cascading effects on the output's characteristics. Some background: dsp.stackexchange.com/questions/54047/what-is-zero-phase-filtering-and-forward-backward-filtering and ruclips.net/video/ue4ba_wXV6A/видео.html
Playing back music is the result of many phase-shift systems in series (from the recording process to sound reproduction. Amir touches on some of them). Barring completely screwed up equipment, this is a complete non issue for the human audio experience, as Amir properly concludes.
Hehe, nice one Amir. Paul might be out of phase with this explanation 😁
Paul as engineer is a great SALESMAN!!
Way I see it, it matters very little because speakers are almost all an absolute nightmare in terms of phase. Minimum phase crossovers rotate phase a lot, and unless you really screw it up, it's not audible. It could be argued that channel-to-channel phase matters some (wouldn't want L and R to be out of phase more than a little bit), but... within a channel? Clearly not important.
The phase shifts introduced by amplifiers are typically small and inaudible compared to phase shifts between multidriver loudspeaker components. A different question is: In an anechoic chamber or in a special listening room where reflections to the listening position are sufficiently absorbed, do loudspeaker phase shifts through the crossover regions of multidriver systems or do "time alignments" of drivers audibly matter ? - if so what time domain accuracy criteria are required for these phase problems to be inaudible?
Paul's truths as he says 😀
Thanks for the video Amir.
My pleasure. Thanks for watching.
I have a question:
- does it mean that traditional speaker crossovers (which introduce phase shifts to the signal) are as good as active crossovers?
What you show here, may show that small harmonic distortions have very minor affect on how we perceive sound when listetnig through speakers
At first when Paul said linearity I thought he was going to say that something advertised as going to 40khz would have a flatter response, which may or may not be valid, but it would at least be audible. Then he said the thing about phase shift and well, I wasn’t impressed. I’m glad you made this video to set the record straight. Nothing against Paul or his viewers, but I rarely watch his videos for anything more than enjoyment. There isn’t much I’ve heard him say aside from very basic topics that is actually proved or well researched.
I think you are too polite to Paul. I'm just a freeman commenter and I don't have to be. No offence, but Paul intentionally lies in many cases. No mistakes nor misunderstandings. He is not a good man with good intentions. He runs a business and he wants to sell equipments, and he knows exactly that the biggest interest comes from goods dabbled with snake-oil. Imagine how easy to manufacture a power cord or an AC filter with a LED. He acts like a kind grampa, but he is far from it. Stay away, he's a charlatan with big ego and confidence. He stated he was an engineer, but I think he lied at that point as well, he's thinking is very far from that.
I find it surprising how good some people are at arguing their beliefs when it comes to science vs phsychology. They're not really dumb as such, just stubborn with beliefs, at least as it appears to us.
You can convince yourself of many things, and overtime, they can become so strong that you would fight to the death for that belief
Amir - I don’t understand why you are so daft. Maybe it’s intentional? “No reference to why phase shift is bad” - didn’t think it was necessary at this level. Swap the wires on your speakers and see what a phase shift does in terms of audio. It’s significant.
A phase shift in frequency domain, group delay in time domain , group delay for all frequency that pass the amplifier should be the same , Paul is right LOL.
Great video Amir. This started me thinking, it would be great to see a technical appraisal of loud speaker designs that are intended to mitigate or even leverage room reflectivity, such as the linkwitz designs.
That's is a hairy and complex project you are proposing. :) I can only do that in the context of reviewing such speakers. The larger project has so many dependencies that even masters of research in this area have punted on it.
@@AudioScienceReview well you either end up a hero with the DIY community or have to go into hiding!😉
Well.. I’ve used a freq variable phase shifter while working on mono recordings and moved forward/back the different frequency bands. Dramatic differences occur for acoustic instruments, especially in transient information but needs to be milliseconds apart, not a hand full of samples to be perceptible. What happens is the spatial cues in the music become more or less coherent. I must have super human hearing based on Amir’s assessment. ; ) Also, speakers are designed to see linear info before they scatter shot sound waves across their modeled listening rooms. How and where the energy propagates might be part of their design choice which is what your are paying for. I’m sure that means something especially in low freq range where cancellation can really suck total net energy from a tonal profile.
You are obviously right. Look up all the patents (eg Dolby) that discuss maintaining phase. Toole misses WHEN phase become very important. Amir would be wise to take his own advice and study more.
@ReaktorLeak That makes senses. Interestingly I could hear changes way more clearly in the 800s HP than the others. There is a go figure!
How do you know your "freq variable phase shifter" didn't also shift the frequency response?
@@AudioScienceReview Not sure, but if it helps config was using a PHA-979 plugin from Voxengo highlighting freq bands I want to move around using band selection in the spectrogram. Of course massive differences on L / R phase shifts but also when tinkering in mono. My goal was to try to shift forward high freq info on a mono piano recording to see if it would brighten things up. Unsuccessful, but spatial cues def were impacted. Would I hear the differences on a speaker system, outside HP, likely no. Could be a cool self test to replicate on your own and peg audible reference point to the shifts. It’ll be higher than spec fueling PS audio’s amp argument for sure ; )
Thank you for all the education Amir! Love your videos.
You did it again my friend, I always learn so much from you, very fascinating
In early CD players, it was incredibly audible with the anti aliasing filters at 22.05 khz.
In loudspeakers it is also very audible.
4th order butterworth crossover adds a full cycle, 360 degree's, of phase shift (time delays) at the crossover point. Thusly, the drivers are NOT starting or stopping at the same time.
How much time delay was imposed? Exactly one full period of the crossover frequency. If that was 400Hz, then the time delay between the two drivers is 1/400th second, or 2.5 milliseconds, which is ~32 inches for time of travel, acoustically on a 4th order network. And that's at the crossover frequency. As you move away from the crossover point(s), the amount of time delay changes at every frequency. In many consumer speakers, the tweeter has started and stopped before the woofer starts. Easily seen in a simple step response test.
If you're telling me this is not audible to you, you're in the wrong line of work.
I also wonder about this.
The woofer and tweeter can be in phase at the crossover frequency on axis, but in fact also be 360 degree off.
I understand that a cloud of reflections from your room mess up the perfect phase coherence, but a short direct sound with the timing between drivers messed up is very audible.
This is the reason one sets time delays between the pieces of large horn soundsystems to make them sound as one thing.
Otherwise it sounds like a bunch of boxes regardless of the frequency response being flat.
Is it just me and Dale?
Dude I’m sorry if my comments are repeatative, but your videos are so epic. Maybe team up with golden boy and demolish what is left of MQA next :p
Any unwanted type of distortion is bad, but the question is how much is there and is it tolerable/preferable and audible.
Phase shift is our best friend. For one thing, it gives us EQ.
I only pre like videos from two creators. Amir and Chef Jean Pierre. These guys are the truth
I tried a tube buffer on my pc and I couldn't localize the direction of the other players around me. If not a phase issue, what then? With another amp that problem went away.
I also use a tube buffer for my desktop, did not have that problem though on headphones or speakers, maybe something else was causing it or the channels / tubes were not balanced properly, (was using a SUCA Audio Tube-T1 with upgraded components inside and power supply also an Xduoo MT-602 with sennheisder HD565s, Philips SHP9500s, koss KPH-30i and others, wharfedale/technics speakers FX-Audio XL-2.1 BL amp, Topping D10S DAC and FX-Audio DAC-X6 with upgraded opamp).
It likely was introducing differential phase shift beetween channels, or other effects. Not what Amir is talking about, i.e. absolute shift equally applied to both channels.
A very simple test of the high-frequency extension needed will be using a 192 kHz sampling frequency recording (or higher) and then insert a very steep low pass filter. This filter can be set at 80 kHz, one can listen to the recording. After this re-listen with the filter set to an ever lower and lower cut off frequency until a perceptible difference between the original recording and the filtered recording becomes audible. If the frequency where the difference becomes audible lies above the frequency the test person/persons can hear with sustained sinus tones it would be interesting (but not absolute proof).
Dr Amar Bose (yes that Bose) wrote ( I have forgotten where so I am quoting from memory) that humans in tests cannot hear the difference between a 8 kHz tone and a 8 kHz square wave. Again not a proof of inaudibility of high frequency phase, but rather an indicator of the lack of importance of extending the frequency response into the stratosphere. Peter Walker (yes the Quad guy) in an interview once said something along the the lines of “the quality of a loudspeaker is determined long before it reaches 10 kHz” (i am so sorry, again I have forgotten where I have read it)
My own take on phase distortion is that it’s inaudible on any well designed loudspeaker with the possible exception of the extreme sub bass where the time delay becomes truly longer with distortion of phase.
Paul was out of Phase. I watch Paul a lot and he does have some inputs from what he knows, like you said it's not always the whole picture or both sides of the story. He's just human, but it's a good idea that if you truly want to help you provide accurate information regardless of the topic.
I like to do more videos like this so if you specific topics like it, please message me on Audio Science Review.
@@AudioScienceReview Honestly Amir I wish I could, I am very new to audio. Less than a year, it's hard for me to correlate fiction and non-fiction, but I am one who is capable of learning and have the intelligence to comprehend and store what I do learn. So I try to watch topics, I dont think anyone does what you are doing on youtube or even through the forums, which I am a member.
I really enjoy and wish I could give you something to look into and correct, but probably someone else can do a better job.
However, I surely would like to see more of these, and honestly better to mix it up rather than make someone feel targeted.
I knew that Paul is a bit iffy especially with his cables talk, he's very adamant about cables, speaker, power, power conditioners. Something I have never agreed with even as being someone who knows nothing about audio and such little experience. I was much more of a follower of him in my early days of audio, now it's so so and his video's are very short which isn't time consuming.
I don't have the right for asking but more videos please.
Point taken however the only real issue with this is you may as well not test anything upstream to the room including the speakers since the end result is dominated by the imperfection of the room/listener interface and yet we know that two loudspeakers can sound different enough to worthy proper measurements. Group delay of a speaker is more worthy a measurement than absolute phase since it tells the difference in time delay versus frequency. I am sure a step response from a first order crossover is far superior to one from a 4th order LR crossover and also audible to the listener.
" you may as well not test anything upstream to the room including the speakers since the end result is dominated by the imperfection of the room/listener interface"
Why did you draw that conclusion from what he said? Where did he state you cannot hear any other differences? He doesn't even imply it yet alone say it.
I show group delay for every headphone I test. I have yet to see any insight from it. It sometimes confirms what we better see in another measurement but that is it.
thank you Amor for this video and for all that you do for the hobby community!
If Im understanding you correctly - the phase/time shifts introduced by EQ is not detectable? Therefore, it would still be okay to use an analogue graphic EQ in a stereo system if one wanted to apply room correction EQ the old fashioned way and there would be no negative effects from a phase shift perspective?
... and the same can be said for passive crossovers - otherwise we'd have even bigger problems anyways?
Phase shift can be used on panel speakers to make a flat panel electrically curved. In this context you can hear the phase shift in what I would consider a positive way. Some designs that do this also work as a low pass filter making the sound off axis sound further away. Some people think further away is depth because of more detail. Interesting how the brain can be tricked in this way.
Most of the time I do agree with you. But do think you should look out with making generalizations.. Reflected sound might be processed differently by our brains than a distorted direct sound. So giving an explanation based on the mess cause by reflected sound might not be relevant. (Know they are both just sounds, but the reflected sound might have some intrinsic properties like a high correlation with sounds recently processed)
Hearing an audio engineer claim that nearly any type of EQ will add audible phase shift to headphones, therefore they need to be hardware tuned for proper tonality or else he deems them unlistenable. Any truth to this? Sounds like an extreme exaggeration and something that just wouldn't be audible to human ears in any real way.
Amir, will you do the blind amplifier test proposed by goldensound?
??? I proposed the blind test, not him. :) But yes, I do plan on performing the test. He ran off with his own idea of how to do one as well.
*I meant the bet he proposed
I demoed a pair of Wilson Audio Alexx speakers in what was probably a $400,000 system. The owner kept on explaining to me how the speakers are ever so slightly angled and at different depths to account for the changes in time domain.
They sounded incredible, don't get me wrong, but i don't think that was the result of "corrected phase and time domain due to the speakers being slightly different distances from my ears" haha
My father who was also an engineer told me many times that if you can't say it with numbers you're full of $hit. So let's compute the phase shift from the graph and see how much change in time there is. An amplifier like all products of this type is called a zero phase device. That means that there is a direct correlation between amplitude frequency response and phase frequency response. It's an exercise for electrical engineering students to compute one, given the other. A loudspeaker is NOT a zero phase system.
In the graph the signal is phase shifted 35 degrees at 20 khz. A full cycle is 360 degrees so this is around 10 percent. The period of a 20 Khz sine wave is 1/20,000 of a second. So a tenth of that is 1/200,000 of a second. That's the time based error. Now I know audiophiles can hear a flea fart in a thunderstorm but I think this is more than a little far fetched.
This correlation between amplitude frequency response and phase frequency response is why you never see a phase response measurement in any amplifier, preamplifier or the like in a review the way you do with speakers where the phase response often looks like a roller coaster. What about an equalizer? Yes when you adjust it, there is a phase difference. But by correcting the amplitude frequency response distortion inherent in the rest of the system (including the recording) you are also correcting the phase frequency response distortion. So this qualitative statement in the video while technically true is so small as to be insignificant. There are lots of problems with sound systems but this is not one of them. Now a certain person claims to be an electrical engineer but hates math. If you've ever sat in classroom in a course in electrical engineering, math is practically all there is and it's not algebra.
It's something that I never understood myself. How can I enjoy live music from a band where the instruments are positioned all over the place and the source of sound is not coming from one place. Yet when it comes to speaker designs we want the tweeters and woofers on a stepped baffle to get them in phase with other. Now unless they are 180 degree out of phase and cancelling each other I can understand but a minor variation in phase shouldn't make a huge difference. Very complicated but very interesting topic. Very well explained by Amir. Thanks.
Great argument Hari. As Dr. Toole says, live presentations are nothing but phase soup!
If (the likes of) Paul wasn't so concerned about his money, ABOVE integrity, he could actually learn from Amir and science, and actually use the information to implement into his designs accordingly, but hey, same old right, there's money in snake oil and he will continue to perpetuate it.
There is so much opportunity for them to build better products instead of chasing voodoo in audio. Somehow they think they know more than entire body of audio science and research to chase these dead ends. Some of their products I have reviewed are fine. Others, they take a concept and just ruin it with these ideas.
@@AudioScienceReview Couldn't agree more Sir.
Also if attenuation starts around 20Khz and we adults don't hear above 15Khz (phase shifted or not) there are really no arguments to be made
The ultimate goal of Paul's u-tube talks is to "blind you with (false) science" and techno-babble, in order to give his company "credibility" to compensate for some of their overpriced "snake-oil" products. I still watch his vids..just for amusement. Occasionally he does give out some accurate "neutral" advice. That said, I would research other vendors BEFORE purchasing ANY PS audio product.
You haven't laughed hard enough until you see Paul with a white board, trying to draw a simple amplifier circuit. Better yet, watch his speaker engineer Chris's face when Paul says something that doesn't compute. Can't contradict the boss.
It was about the same when Guttenberg interviewed the young cartridge engineer, Parker. The subject became cleaning vinyl. When Steve said he just washed his LP's in the kitchen sink with dish detergent, Parker had to look away and wince. His long silence was palpable. Credit is due: he didn't double over laughing.
All that said, you can't expect to make lots of followers by goring sacred YT bulls.
@ReaktorLeak I tried it once. It leaves a bunch of residue -- both minerals from tap water and chemistry in the detergent, I found. It's why commercial disc washers use distilled water with specialized cleaning solution. Today, I use an Audio Technica brush and cleaning solution with great success.
@@jimshaw899 I might do this for a really filthy record before cleaning with an RCM.
I get the impression this is only in the context of amplifier phase shift that is both too short and too high frequency to be audible? What about in the context of speakers where drivers may have different relative delays?
I have monitors with DSP to undo their inherent phase shift, as measured very precisely by the manufacturer and it makes a definitely audible difference, although I have to listen in near-field (i.e. direct sound is far higher than any reflections) to get that. I think a growing number of active monitors have this feature built in. It's quite interesting to turn it on and realise just how tight certain musicians can play together, or when someone in editing has shifted the notes to perfectly fit a grid.
Also on the subject of these small speaker-dependent delays, I had a thought on depth of 'soundstage' - an important consideration for many people buying speakers and headphones. Could having quite highly frequency-dependent phase delay actually artificially cause the illusion of depth? I suggest that since every 15 or so milliseconds of relative delay would be equivalent to an extra five metres distance to the performer. Pick your delays right and perhaps you can mix individual's recordings to have the soundstage of an orchestra?
Phase is extremely important when designing speakers though, (if they are two-way or more). Now I wonder about designing the drivers themselves, there must be some thinking to be done about phase too, right?
Relative phase matters at crossover frequencies. The total phase shift hardly matters in realistic applications.
@@hidjedewitje I was just confused for a moment about the title: "Does Phase Distortion/Shift Matter in Audio? (no*)" Because it depends on the context you are talking about.
@@TobyIKanoby Well, it's all about how you phrase it and what is a phase shift. If you have 1 speaker that plays a sine for a second and then a -sine for a second. You won't hear a difference despite the worst case scenario of 180 degrees phase shift.
If you have two sources playing at the same time, you have 1 playing the sine and the other playing the minus sine, you will hear something different than when both sources play the same sine. This is because sine +-sine=0 and sine+sine=2sine and these are obviously not equal. This is exactly the case in loudspeakers at the crossover frequency! Should we now say phase shift is audible because this experiment suggests so? No because in this experiment the amplitude and dispersion also change! Those parameters are very audible!
There is also phase modulation. This happens when a sine changes it's phase depending on another signal. This is very non-linear and will thus cause extra tones to appear in the frequency spectrum (also audible!).
@@TobyIKanoby That is what the asterisk (*) was for. :) Of course phase is an important electrical characteristic in many topics. It is just that it doesn't matter in the way most people like Paul try to sell you something else. As I mentioned, if you know the context, then sure, pay attention to phase. But if you don't, don't blindly assume it is something important.
@@AudioScienceReview as I said: I was a little bit confused by the title first until you explained the context, as you always do very well, thanks.
Funny thing is I actually saw the original video and didn't really understand what he was talking about so decided to just ignore, I can't hear much above 14k anyway. Now you explained what he was talking about and why my reaction to ignore it was the right one, great :). As I said: thanks!
To see impact of all phase distortions the simulation and frequency sweep are useless 100% Soft linear shifts are marginal and due to no much comparable time of impulse do not affect amplitude of outcome That is what you measured
But to see impact of phase shifts for realty of listening one should consider speaker signal and it's reflection from any surrounding surface (which will provide time delay) and analyze summary of both at the same time . In that situation even for normal level signal fading to zero is easy noticed in many bands
To prove it please place your open hand at side of tweeter (you make this way source of delayed signal) and listen to noise and see how timbre is affected Frequency sweep is not happening at the same time thus any deductions or summary of signals cannot happen and this sweep results are totally not true in compare to listening. Ear is not a testing bench and do not sample music by each harmonic :).
Wouldn't there be differential phase between frequencies though?
The phase shifting at speaker crossovers is taken into account in order to avoid a phase cancelling notch in the frequency response.
This can be measured.
A pure Butterworth crossover for example would require the tweeter to be wired the opposite way round to the woofer for smooth integration.
Does this come under differential phase effects which should continue to be observed?
At this point in time, all passive crossover types that are a sensible solution as in parts count,price and driver roll off , have been studied to death and the results are well known.
For all the speakers being sold out there, they all work well IF designed and measured properly and they all have a variance in type of crossover used. Yes, the phase is all over the place but who can hear the difference if the frequency response is optimal?
@@Martin-eelab And your point is?
@@Martin-eelab I already addressed the crossover question and so did Amir in the video. It's a non issue according to Floyd Toole and others. I don't get what you are trying to say honestly. Amir showed how a room swamps out the nice neat speaker response with reflections.
@@Martin-eelab Of course reversing the tweeter polarity causes an amplitude difference. You are messing up the combined crossover response between the two drivers. If the crossover is designed to have flat or near flat response , then reversing the polarity of a driver of any type of crossover will change the amplitude. So I see this being a discussion of so what? Leave things alone as designed.
Best video ever that sheds light on many misconceptions the HI-fi industry comes up with.Making a little joke...the phase distortion\shift after 20k only makes the bats disorientated🤣🤣
This is by far the most objective channel on audio equipment I've seen on YT but in this case the qualifying statements in the cited text would make 'subtle effect' and 'when it has been audible' a tiny opening for Pauls claims to be treated a little more positively I would think; especially if you've got an expensive treated room to match the expensive equipment you put in it. Having said that, in my experience, high power audio amplifiers are not too stable into all speaker loads if they're allowed to produce frequencies as high as 100KHz. Filtering out frequencies above 20KHz is more of a practical choice for stable product design than worrying about phase changes in frequencies we can't hear and never existed in the original recording.
Paul may know how to sell audio, but he knows very little about it. You would think in 30 or 40 years, he would learn something about how the auditory system works, but that appears beyond him. When looking at this problem, we need to look at how we hear. Except in a narrow frequency band we don't have any concept of high resolution timing, and even over that band, it is stereo differential timing, not absolute timing. Our auditory receptors are not narrow band, so rapid phase changes w.r.t. frequency have a mechanism for auditory impact, but slow ones do not.
If it is possible to hear phase distortion under any circumstances then audiophiles are going to be concerned about it. If we say it’s inaudible under normal conditions then that’s simply explained by low resolving systems or deaf listeners.
Resolution of your system doesn't matter. Your room doesn't come in high resolution or low resolution. :) It creates massive phase distortion on its own that nothing you can do up stream eliminates it. It is outside of the control of the hardware you use.
@@AudioScienceReview That explanation works for me, but the fact that phase distortion is audible under special conditions (however contrived) will be sufficient encouragement for audiophiles who believe they can hear the phase distortion of high-order low-pass filtering the 20 kHz anti-aliasing or reconstruction filters used in standard CD audio. I think we have a significant indication of the relative unimportance of phase distortion when we consider that some highly-regarded loudspeakers use 4th order filters at 300 Hz or 3 kHz, where ears are naturally most sensitive.
How much reflection do you get inside the ear canal using IEMs
Good question. It is substantially less but there is still some in the ear canal. See my review of Panasonic RP-TCM125-A: www.audiosciencereview.com/forum/index.php?threads/panasonic-rp-tcm125-review-budget-iem.22797/
You can still see some cancellation occurring around 8 kHz which has to be caused by reflection.
Finally, someone has came along to callout all the BS spreaders!
You simply cannot get true linear phase.
So paying to much attention to phase shift that happens on a very subtle level is kind of funny because you get a much much much bigger problem: your room.
You re missing the point. Just because small phase shifts do not make a whole lot of difference does not mean that it should be ignored. After all, audiophile equipment is all about trying to achieve perfection.
Sounds like you are missing the point when you realize that your room is a phase soup. As is any live presentation of music. Mind you, phase perfection is fine as long as it doesn't come with its own problems. Companies like Wilson attempt to compensate for phase by having each driver in its own box and different distance to listener. Problem is, how they have localized reflections and diffraction. Crossover designers make similar mistake by using low order ones to avoid phase shifts and with it, screw up the tonality which you can hear. In general, phase accuracy is synonymous with bad engineering in audio and is uses as a marketing tool.
Before I encountered ASR, Paul's channel was one of many (GR research anyone?) the YT algorithm pushed on me. How young, ignorant and impressionable I was:)
But Paul never convinced me of anything else than his dubious sales techniques. His explanations are glorified wordsalad.
Great video Amir glad you're creating videos.
Thanks and appreciate you watching.
Did amir get into trouble with paul ps audio with trashing his snake oil with measurements cos not seen an amir video for over a month now
Great insight Amir, thanks. I've always had an interest in the effect of phase changes across the frequency band. I was interested in the result for the closed back headphones and the bouncing around of reflections within the cup, acting as a "mini room" and so suffering from the same effects as a room. What would be the case with open backed headphones like I have (Grado 225's). If im sat in a quiet room, will I still have reflections going on around my ear if the headphone are of the open type? Would the unwanted, indirect sounds just escape into the room and not be reflected back towards my ear thus avoiding the "phase soup" effect?
You would have to be very close to a wall for the reflections to matter. I am thinking a foot or less. So in general this is not a worry for headphone listeners.
Difference between salesman and engineer!
Thanks for yet another valuable lesson Amir.
Main issue is measurement but maybe you don’t know what measurements need? Or silver sounds same like gold or other metals? If you cannot answer to yourself why difference in sound may be need to start search ? Or if you cannot hear may be need to ask you why ? May be problem with you ? Many questions
Great video and explanation Amir. Excuse my ignorance, but does this mean that the 0 to 180 phase switch on the back of subwoofers is irrelevant or unnecessary? Side question, but does this mean phase rotation is irrelevant? Thank you for your videos Amir!
No it doesn't mean that. That would come under differential phase, which he talks about, albeit only in reference to left and right ears, but it also applies to additional speakers.